Diagnostic Performance of Clinical Laboratory Indicators With Sarcopenia: Results From the West China Health and Aging Trend Study

The original scores of traditional Chinese medicine constitutions are risk and diagnostic factors in middle-aged and older adults with sarcopenia

Objective

Sarcopenia is a geriatric syndrome that occurs with age and is characterized by a gradual decline in muscle mass, power, and functionality. It serves as a prominent contributor to frailty, disability, and mortality among older individuals. Currently, no standardized global guidelines exist for the diagnosis of sarcopenia. This study aimed to establish the correlation between sarcopenia and the constitutions of traditional Chinese medicine (TCM), considering the connection between physical functioning and sarcopenia.

Methods

A total of 1441 participants in this study were diagnosed with sarcopenia. The Asian Working Group for Sarcopenia (AWGS) proposed a sarcopenia definition algorithm. To determine the constitution of each participant, classification and determination standards were used in traditional Chinese medicine. This study evaluated the demographics, lifestyles, and self-reported medical history of individuals diagnosed with sarcopenia through a self-administered questionnaire. The constitution of the participants was determined using TCM classification and determination standards. Subsequently, we analyzed the results of univariate analysis and multivariate regression and constructed a receiver operating characteristic (ROC) curve.

Results

Participants who were diagnosed with sarcopenia had substantially lower original Neutral constitution scores (P < 0.050). In comparison to those without sarcopenia, individuals with sarcopenia exhibited notably elevated original Qi-deficiency, Yang-deficiency, Yin-deficiency, Blood-stagnation, and Qi-stagnation scores in contrast to those in the healthy group (P < 0.050). The identified risk factors associated with sarcopenia included the following: Neutral (OR = 0.903), Qi-deficiency (in males, OR = 1.126), Yang-deficiency (OR = 1.062), Phlegm-dampness (in males, OR = 0.833), and Blood-stagnation (in females, OR = 1.089). The highest area under the curve (AUC) was observed for the original neutral constitution score, followed by the Yang-deficiency and blood-stagnation scores (0.644, 0.613, and 0.611, respectively). Additionally, the AUC for the combined original scores of all nine constitutions among males reached 0.778.

Conclusions

In this cross-sectional study of older people with higher original Qi-deficiency, Yin deficiency, Yang-deficiency, Blood-stagnation, and Qi-stagnation were associated with sarcopenia. Notably, various TCM constitutions are significantly linked to sarcopenia. There was a significant occurrence of various body constitution types among individuals diagnosed with sarcopenia. The mixture of the nine original constitution scores exhibited good diagnostic performance for sarcopenia in males.

Neoadjuvant therapy increases the risk of metabolic disorders and osteosarcopenia in patients with early breast cancer

BACKGROUND

The purpose of this study is to evaluate the effects of neoadjuvant therapy on glucose and lipid metabolism, bone mineral density (BMD) and muscle, and to explore the relationship between metabolic disorders and changes in body composition, so as to provide better health management strategies for breast cancer survivors.

METHODS

The clinical data of 43 patients with breast cancer who received neoadjuvant therapy in Xuanwu Hospital from January 2020 to June 2021 were analyzed retrospectively. The biochemical results, including albumin, blood glucose, triglyceride and cholesterol, were collected before neoadjuvant therapy and before surgery. The pectoral muscle area, pectoral muscle density and cancellous bone mineral density of the 12th thoracic vertebra were also measured by chest CT.

RESULTS

After neoadjuvant therapy, fasting blood glucose, triglyceride and cholesterol were significantly increased, albumin was decreased. At the same time, pectoral muscle area, pectoral muscle density and T12 BMD were decreased. After treatment, BMD was positively correlated with pectoral muscle area, R2 = 0.319, P = 0.037, and BMD was also positively correlated with pectoral muscle density, R2 = 0.329, P = 0.031. Multivariate analysis showed that BMD and pectoral muscle density were correlated with menstrual status, and pectoral muscle area was correlated with body mass index before treatment, none of which was related to glucose and lipid metabolism.

CONCLUSION

Neoadjuvant therapy can cause glucose and lipid metabolism disorder, BMD decrease and muscle reduction. BMD was positively correlated with muscle area and density after treatment, suggesting that patients had an increased chance of developing osteosarcopenia.

Association of High-Density Lipoprotein Cholesterol with Sarcopenia in Chinese Community-Dwelling Middle-Aged and Older Adults: Evidence from 4-Year Longitudinal Study

INTRODUCTION

There is inconsistent evidence on the associations between high-density lipoprotein cholesterol (HDL-C) and risk of sarcopenia. The aim of the study was to determine the evidence existing between HDL-C and sarcopenia in Chinese middle-aged and older adults.

METHODS

We used a panel study design of the China Health and Retirement Longitudinal Study (CHARLS), with 7,415 participants (mean age 57.5 years) from 2011, 2013, and 2015. HDL-C was measured by colorimetric test of venous serum samples. Sarcopenia was defined as low muscle mass, plus low muscle strength, or low physical performance. Muscle mass was estimated by anthropometric measures. Muscle strength was measured by handgrip strength using dynamometer. Physical performance was measured by 5-time chair stand test, gait speed test, and short physical performance battery.

RESULTS

With 961 (13.0%) sarcopenia cases, each 1-unit increase (1 SD = 15.4 mg/dL) of HDL-C levels was associated with 42% increased odds of incident sarcopenia (OR = 1.42, 95% confidence interval [CI] = 1.28-1.58) at 4-year follow-up. Females with high HDL-C levels (HDL-C >60 mg/dL) had a higher risk of sarcopenia (OR = 2.49, 95% CI = 1.76-3.52). The restricted cubic spline curves showed a J-shaped association between HDL-C and risk of sarcopenia in females. HDL-C was negatively associated with muscle mass (β = -0.23, 95% CI = -0.27 to -0.20) and hand grip strength (β = -0.05, 95% CI = -0.19 to 0.09).

CONCLUSION

High HDL-C levels were associated with higher risk of sarcopenia among middle-aged and older Chinese adults, and appropriate control of its high levels informs the management of sarcopenia.

The diagnostic performance of Cr/CysC for sarcopenia and its predictive value on clinical outcomes in hospitalized older patients: a prospective cohort study

PURPOSE

The utilization of the creatinine-to-cystatin C ratio (Cr/CysC) represents an innovative method for predicting sarcopenia. Our objectives encompassed the evaluation of sarcopenia diagnostic accuracy for Cr/CysC, SARC-F, SARC-CalF, the combination of Cr/CysC and SARC-CalF, and the Ishii score, as well as an exploration of the predictive value of Cr/CysC concerning clinical outcomes within hospitalized older individuals.

METHODS

We employed receiver operating characteristic (ROC) curves and calculated areas under the curves (AUCs) to assess the diagnostic accuracy. Furthermore, we applied univariate and multivariate Cox proportional-hazard models to calculate the hazard ratio (HR) and 95% confidence interval (CI) of risk factors affecting prognosis.

RESULTS

Our study included 312 participants, comprising 167 men and 145 women, with an average age of 71 years. Among males, the AUCs for Cr/CysC, SARC-F, SARC-CalF, the combination of Cr/CysC and SARC-CalF, and the Ishii score were 0.717 [95% CI 0.642-0.784], 0.669 (95% CI 0.592-0.739), 0.845 (95% CI 0.781-0.896), 0.882 (95% CI 0.823-0.926), and 0.938 (95% CI 0.890-0.969), respectively. In females, the AUCs for Cr/CysC, SARC-F, SARC-CalF, the combination of Cr/CysC and SARC-CalF, and the Ishii score were 0.706 (95% CI 0.625-0.779), 0.631 (95% CI 0.547-0.710), 0.763 (95% CI 0.686-0.830), 0.789 (95% CI 0.714-0.853), and 0.898 (95% CI 0.837-0.942), respectively. After adjusting for age, sex, physical exercise, smoking, drinking, hypertension, coronary heart disease (CHD), chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), and cancer, sarcopenia identified by Cr/CysC (adjusted HR = 2.176, 95% CI 1.062-4.460, P = 0.034) was independently associated with poor overall survival in hospitalized older patients.

CONCLUSIONS

Cr/CysC has satisfactory diagnostic accuracy for sarcopenia diagnosis and predictive value for poor outcomes in hospitalized older patients. The combination of Cr/CysC and SARC-CalF may provide a more accurate screening for sarcopenia and the Ishii score may be the most accurate clinical method for detecting sarcopenia.

The relationship between sarcopenia and metabolic dysfunction-associated fatty liver disease among the young and middle-aged populations

BACKGROUND

Metabolic dysfunction-associated fatty liver disease (MAFLD) has been proposed as a new term for diagnosing fatty liver disease, which is considered to be a multi-systemic disease with multiple extrahepatic manifestations, including sarcopenia. The link between sarcopenia and MAFLD remains uncertain, especially among young and middle-aged adults. Thus, we examined the relationship between MAFLD and sarcopenia in young and middle-aged individuals in this study.

METHODS

A total of 2214 individuals with laboratory tests, dual-energy X-ray absorptiometry and ultrasound transient elastography from NHANES 2017-2018 were selected for this study. MAFLD was diagnosed as fatty liver disease with any one of the situations: overweight/obesity, diabetes mellitus, presence of metabolic dysregulation. Sarcopenia was defined by appendicular lean mass adjusted for body mass index (BMI). Multivariable logistic regression and restricted cubic spline (RCS) model were applied to explore the relationship between MAFLD and sarcopenia, and the mediation analyses were also conducted. Moreover, subgroup analyses stratified by BMI and lifestyles were done.

RESULTS

The prevalence of MAFLD was 47.85%, and nearly 8.05% of participants had sarcopenia. The prevalence of sarcopenia was higher in participants with MAFLD (12.75%; 95% CI 10.18-15.31%) than in the non-MAFLD (3.73%; 95% CI 2.16-5.31%). MAFLD was significantly positively associated with sarcopenia after adjustments [OR = 2.87 (95% CI: 1.62-5.09)]. Moreover, significant positive associations were observed between liver fibrosis and sarcopenia prevalence in MAFLD patients (OR = 2.16; 95% CI 1.13-4.15). The RCS curve revealed that MAFLD was linearly associated with sarcopenia. The relationship between the MAFLD and sarcopenia were mediated by C-reactive protein (mediation proportion: 15.9%) and high-density lipoprotein cholesterol (mediation proportion: 18.9%). Subgroup analyses confirmed the association between MAFLD and sarcopenia differed in different lifestyle groups.

CONCLUSIONS

Both MAFLD prevalence and severity was significantly associated with sarcopenia. Thus, clinicians should advise comorbidity screening and lifestyle changes to young and middle-aged patients.

Blood biomarkers for sarcopenia: A systematic review and meta-analysis of diagnostic test accuracy studies

Biomarkers are emerging as a potential tool for screening or diagnosing sarcopenia. We aimed to summarize the current evidence on the diagnostic test accuracy of biomarkers for sarcopenia. We comprehensively searched Ovid MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials up to January 2023 and only included diagnostic test accuracy studies. We identified 32 studies with 23,840 participants (women, 58.26%) that assessed a total of 30 biomarkers. The serum creatinine to cystatin C ratio (Cr/CysC) demonstrated a pooled sensitivity ranging from 51% (95% confidence interval [CI] 44-59%) to 86% (95% CI 70-95%) and a pooled specificity ranged from 55% (95% CI 38-70%) to 76% (95% CI 63-86%) for diagnosing sarcopenia defined by five different diagnostic criteria (11 studies, 7240 participants). The aspartate aminotransferase to alanine aminotransferase ratio demonstrated a pooled sensitivity of 62% (95% CI 56-67%) and a pooled specificity of 66% (95% CI 60-72%) (3 studies, 11,146 participants). The other 28 blood biomarkers exhibited low-to-moderate diagnostic accuracy for sarcopenia regardless of the reference standards. In conclusion, none of these biomarkers are optimal for screening or diagnosing sarcopenia. Well-designed studies are needed to explore and validate novel biomarkers for sarcopenia.

Relationship between lipid profiles and reduced handgrip strength (dynapenia) in hospitalized patients with cirrhosis

BACKGROUND

Dynapenia embraces clinical significance and predictive value separated from skeletal muscle loss among cirrhosis. Moreover, alterations in lipid levels may impact muscle function. It has yet to elucidate the relationship between lipid profiles and muscle strength weakness. We sought to explore which lipid metabolism indicator could be useful to identify patients with dynapenia in daily practice.

METHODS

A retrospective observational cohort study enrolling 262 cirrhotic patients. Analysis of the receiver operating characteristic (ROC) curve was performed to determine the discriminatory cutoff for dynapenia. Multivariate logistic regression was conducted to assess the association between total cholesterol (TC) and dynapenia. Also, we established a model based on the classification and regression tree method.

RESULTS

ROC implicated a cutoff of TC ≤ 3.37 mmol/L to identify dynapenia. Patients with TC ≤ 3.37 mmol/L showed significantly lower handgrip strength (HGS; 20.0 vs. 24.7 kg, P  = 0.003), lower hemoglobin, lower platelet, lower white blood cell count, lower sodium and higher prothrombin-international normalized ratio. A positive correlation was found between TC and HGS values ( r  = 0.1860, P  = 0.003). TC remained a significant association with dynapenia after controlling for variables including age, sex, BMI, and the presence of ascites. The decision tree incorporating TC, BMI, and age had a sensitivity of 71.4%, specificity of 64.9%, and an area under ROC of 0.681.

CONCLUSION

TC ≤ 3.37 mmol/L was significantly associated with the presence of dynapenia. Assessing TC may be helpful for identifying dynapenic patients with cirrhosis in the health care or hospital setting.

Machine and deep learning-based clinical characteristics and laboratory markers for the prediction of sarcopenia

BACKGROUND

Sarcopenia is an age-related progressive skeletal muscle disorder involving the loss of muscle mass or strength and physiological function. Efficient and precise AI algorithms may play a significant role in the diagnosis of sarcopenia. In this study, we aimed to develop a machine learning model for sarcopenia diagnosis using clinical characteristics and laboratory indicators of aging cohorts.

METHODS

We developed models of sarcopenia using the baseline data from the West China Health and Aging Trend (WCHAT) study. For external validation, we used the Xiamen Aging Trend (XMAT) cohort. We compared the support vector machine (SVM), random forest (RF), eXtreme Gradient Boosting (XGB), and Wide and Deep (W&D) models. The area under the receiver operating curve (AUC) and accuracy (ACC) were used to evaluate the diagnostic efficiency of the models.

RESULTS

The WCHAT cohort, which included a total of 4057 participants for the training and testing datasets, and the XMAT cohort, which consisted of 553 participants for the external validation dataset, were enrolled in this study. Among the four models, W&D had the best performance (AUC = 0.916 ± 0.006, ACC = 0.882 ± 0.006), followed by SVM (AUC =0.907 ± 0.004, ACC = 0.877 ± 0.006), XGB (AUC = 0.877 ± 0.005, ACC = 0.868 ± 0.005), and RF (AUC = 0.843 ± 0.031, ACC = 0.836 ± 0.024) in the training dataset. Meanwhile, in the testing dataset, the diagnostic efficiency of the models from large to small was W&D (AUC = 0.881, ACC = 0.862), XGB (AUC = 0.858, ACC = 0.861), RF (AUC = 0.843, ACC = 0.836), and SVM (AUC = 0.829, ACC = 0.857). In the external validation dataset, the performance of W&D (AUC = 0.970, ACC = 0.911) was the best among the four models, followed by RF (AUC = 0.830, ACC = 0.769), SVM (AUC = 0.766, ACC = 0.738), and XGB (AUC = 0.722, ACC = 0.749).

CONCLUSIONS

The W&D model not only had excellent diagnostic performance for sarcopenia but also showed good economic efficiency and timeliness. It could be widely used in primary health care institutions or developing areas with an aging population.

TRIAL REGISTRATION

Chictr.org, ChiCTR 1800018895.

Using machine learning to detect sarcopenia from electronic health records

Introduction

Sarcopenia (low muscle mass and strength) causes dysmobility and loss of independence. Sarcopenia is often not directly coded or described in electronic health records (EHR). The objective was to improve sarcopenia detection using structured data from EHR.

Methods

Adults undergoing musculoskeletal testing (December 2017-March 2020) were classified as meeting sarcopenia thresholds for 0 (controls), ≥1 (Sarcopenia-1), or ≥2 (Sarcopenia-2) tests. Electronic health record diagnoses, medications, and laboratory testing were extracted from the Indiana Network for Patient Care. Five machine learning models were applied to EHR data for predicting sarcopenia.

Results

Of 1304 participants, 1055 were controls, 249 met Sarcopenia-1 and 76 met Sarcopenia-2. Sarcopenic participants were older, with higher fat mass, Charlson Comorbidity Index, and more chronic diseases. All models performed better for Sarcopenia-2 than Sarcopenia-1. The top performing models for Sarcopenia-1 were Logistic Regression [area under the curve (AUC) 71.59 (95% confidence interval [CI], 71.51-71.66)] and Multi-Layer Perceptron [AUC 71.48 (95%CI, 71.00-71.97)]. The top performing models for Sarcopenia-2 were Logistic Regression [AUC 91.44 (95%CI, 91.28-91.60)] and Support Vector Machine [AUC 90.81 (95%CI, 88.41-93.20)]. For the best Logistic Regression Model, important sarcopenia predictors included diabetes mellitus, digestive system complaints, signs and symptoms involving the nervous, musculoskeletal and respiratory systems, metabolic disorders, and kidney or urinary tract disorders. Opioids, corticosteroids, and antihyperlipidemic drugs were also more common among sarcopenic participants.

Conclusions

Applying machine learning models, sarcopenia can be predicted from structured data in EHR, which may be developed through future studies to facilitate large-scale early detection and intervention in clinical populations.

Serum transthyretin and aminotransferases are associated with lean mass in people with coronary heart disease: Further insights from the CARE-CR study

Background

Low muscle mass disproportionately affects people with coronary heart disease compared to healthy controls but is under-researched and insufficiently treated. Inflammation, poor nutrition, and neural decline might contribute to low muscle mass. This study aimed to assess circulatory biomarkers related to these mechanisms [albumin, transthyretin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and C-terminal agrin fragment] and their relationship with muscle mass in people with coronary heart disease. Our findings could be beneficial to indicate mechanisms of sarcopenia, detect sarcopenia, and evaluate treatment.

Methods

Serum blood samples from people with coronary heart disease were analysed for biomarker concentrations using enzyme-linked immunosorbent assays. Skeletal muscle mass was estimated using dual X-ray absorptiometry derived appendicular lean mass and reported as skeletal muscle index (SMI; kg m^-2), and as a proportion of total body mass [appendicular skeletal mass (ASM%)]. Low muscle mass was defined as a SMI <7.0 and <6.0 kg m^-2, or ASM% <25.72 and <19.43% for men and women, respectively. Associations between biomarkers and lean mass were adjusted for age and inflammation.

Results

Sixty-four people were assessed; 14 (21.9%) had low muscle mass. People with low muscle mass had lower transthyretin (effect size 0.34, p = 0.007), ALT (effect size 0.34, p = 0.008), and AST (effect size 0.26, p = 0.037) concentrations, compared to those with normal muscle mass. SMI was associated with inflammation-corrected ALT (r = 0.261, p = 0.039) and with inflammation- and age-adjusted AST/ALT ratio (r = -0.257, p = 0.044). Albumin and C-terminal agrin fragment were not associated with muscle mass indices.

Conclusion

Circulatory transthyretin, ALT and AST were associated with low muscle mass in people with coronary heart disease. Low concentrations of these biomarkers might indicate that low muscle mass is partially explained by poor nutrition and high inflammation in this cohort. Targeted treatments to address these factors could be considered for people with coronary heart disease.

Determination of skeletal muscle mass by aspartate aminotransferase / alanine aminotransferase ratio, insulin and FSH in Chinese women with sarcopenia

BACKGROUND

Sarcopenia is an age-related skeletal muscle disorder that involves a loss of muscle mass or strength and physiological function. Skeletal muscle deteriorates in both quantity and quality. The endocrine system is an important regulator of muscle metabolism. Therefore, we aimed to explore the relationship between biochemical markers and muscle mass in sarcopenia.

METHODS

We used the AWGS 2014 as the diagnostic criteria for sarcopenia, considering both the loss in muscle mass, muscle strength and physical performance. A total of 2837 elderly female participants over 50 years of age from the West China Health and Aging Trend (WCHAT) study were included. Insulin, glucose, 25(OH)VD, procalcitonin, alanine aminotransferase, aspartate aminotransferase, total protein, prealbumin, albumin, thyroid-stimulating hormone, free triiodothyronine, free tetraiodothyronine, triglycerides, cholesterol, high-density lipoprotein, very low-density lipoprotein, cortisol, and follicle-stimulating hormone were measured. Based on the findings of univariate analysis, multivariate regression and receiver operating characteristic (ROC) curves were established.

RESULTS

Participants with sarcopenia had significantly lower free triiodothyronine, insulin, total protein, albumin, prealbumin, albumin/prealbumin ratio (A/G), alanine aminotransferase, triglycerides, and very low-density lipoprotein concentrations (P < 0.05). Compared with those without sarcopenia, those with sarcopenia had significantly higher free tetraiodothyronine, cortisol, follicle-stimulating hormone (FSH), aspartate aminotransferase/alanine aminotransferase ratio (AST/ALT), and high-density lipoprotein concentrations (P < 0.05). Insulin (OR = 0.854), FSH (OR = 1.016), and the AST/ALT ratio (OR = 1.819) were independent risk factors for low muscle mass (P < 0.001). The AUC of insulin was the highest, followed by the AST/ALT ratio and FSH (0.691, 0.671, and 0.634, respectively), and the AUC of the mixture of the above three reached 0.736.

CONCLUSION

In this cross-sectional study of elderly Chinese females aged over 50 years from the WCHAT, FSH, insulin, and AST/ALT ratio were associated with sarcopenia and risk factors for low muscle mass.

Clinical muscle mass-related biomarkers that predict mortality in older patients with community-acquired pneumonia

Objective

Community-acquired pneumonia (CAP) is associated with elevated morbidity and mortality, and it usually occurs in older adults. Our goal here was to assess the efficacies of muscle mass-related biomarkers, such as, aspartate transaminase/alanine transaminase (AST/ALT) and creatinine/cystatin C*100 (Cr/CysC*100), in predicting 1-, 2-, and 3-year mortalities of older CAP patients.

Methods

Design: Retrospective cohort study. Setting and Participants: A teaching hospital in western China. Hospitalized CAP patients, aged≥60 years. We separated patients into a high or low muscle mass group, according to the median AST/ALT and Cr/CysC*100, respectively. We acquired data from medical records and local government mortality databases, as well as telephonic interviews. We analyzed the association between low muscle mass (AST/ALT and Cr/CysC*100) and all-cause mortality at 1, 2, and 3 years in older patients with CAP.

Results

We enrolled 606 patients (58.58% male; median age: 81 years) for analysis. The 1-, 2-, and 3-year mortality in older patients with CAP in the low muscle mass group (AST/ALT) was higher than in the high muscle mass group (AST/ALT) (1-year: 51.16% vs. 36.96%, p < 0.001; 2-year: 54.46% vs. 41.25%, p = 0.001; 3-year: 54.79% vs. 42.9%, p = 0.003). Upon adjustment of potential confounding factors, we revealed, using cox regression analysis, that the low muscle mass group (AST/ALT) experienced enhanced mortality risk at the 1-, 2-, and 3-year follow-ups, compared to the high muscle mass group (AST/ALT) (1-year: hazard ratios (HR) = 1.46, 95% confidence interval (CI): 1.13–1.88; 2-year: HR = 1.39, 95% CI: 1.09–1.77; 3-year: HR = 1.35, 95% CI: 1.06–1.72). The 1-, 2-, and 3-year mortality of older CAP patients in the low muscle mass group (Cr/CysC*100) was also higher than the high muscle mass group (Cr/CysC*100) (1-year: 56.29% vs. 31.91%, p < 0.001; 2-year: 60.26% vs. 35.53%, p < 0.001; 3-year: 61.26% vs. 36.51%, p < 0.001). Compared to the high muscle mass group (Cr/CysC*100), the low muscle mass group (Cr/CysC*100) experienced enhanced mortality risk at the 1-, 2-, and 3-year follow ups (1-year: HR = 1.9, 95% CI: 1.46–2.48; 2-year: HR = 1.85, 95% CI: 1.44–2.39; 3-year: HR = 1.85, 95% CI: 1.44–2.37).

Conclusions

Low muscle mass (AST/ALT and Cr/CysC*100) were associated with enhanced 1-, 2-, and 3-year mortality risk in older patients with CAP.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-022-03626-y.